Professional Documents
Culture Documents
July 2009
REVERSE
MORTGAGES
GAO-09-836
July 2009
REVERSE MORTGAGES
Accountability Integrity Reliability
Highlights
Highlights of GAO-09-836, a report to
Policy Changes Have Had Mostly Positive Effects on
Lenders and Borrowers, but These Changes and
Market Developments Have Increased HUD's Risk
congressional committees
Letter 1
Background 3
Most HECM Lenders View the Overall Effect of the HERA
Provisions as Neutral or Positive for Their Reverse Mortgage
Business 11
HERA Provisions Will Affect Borrower Costs and Loan Amounts
Differently Depending on Home Value and Other Factors 20
HUD Has Enhanced Its Analysis of HECM Program Costs but
Changes in House Price Trends and Higher Loan Limits Have
Increased HUD’s Risk of Losses 27
Agency Comments and Our Evaluation 37
Tables
Table 1: Change in Up-front Costs for $300,000 Houses in Various
Locations 21
Table 2: Survey Population and Sample Dispositions 39
Table 3: Universe of 2007 HECMs by Home Value and Maximum
Claim Amount Category 43
Figures
Figure 1: Comparison of 30-Year Forward and Reverse Mortgages 4
This is a work of the U.S. government and is not subject to copyright protection in the
United States. The published product may be reproduced and distributed in its entirety
without further permission from GAO. However, because this work may contain
copyrighted images or other material, permission from the copyright holder may be
necessary if you wish to reproduce this material separately.
A reverse mortgage is a loan that converts the borrower’s home equity into
payments from a lender and typically does not require any repayment as
long as the borrower continues to live in the home. Available to
homeowners aged 62 and older, these loans have become an increasingly
popular financial tool for seniors. Almost all reverse mortgages are
currently made under the Home Equity Conversion Mortgage (HECM)
program administered by the Federal Housing Administration (FHA) of the
Department of Housing and Urban Development (HUD). 1 FHA insures
lenders against losses on these mortgages, and charges borrowers
insurance premiums to cover anticipated insurance claims. The number of
HECMs made has grown rapidly in recent years, rising from 157 loans in
fiscal year 1990 to more than 112,000 loans in fiscal year 2008.
The Housing and Economic Recovery Act of 2008 (HERA) made several
modifications to the HECM program. 2 The first of these changes affects
the origination fees that borrowers pay for these loans. Prior to HERA, the
origination fee was 2 percent of the “maximum claim amount”—the lesser
of the home value or the HECM loan limit. HERA changed the fee
calculation to 2 percent of the maximum claim amount up to $200,000 plus
1
For information on consumer protection issues regarding HECMs, see GAO, Product
Complexity and Consumer Protection Issues Underscore Need for Improved Controls
over Counseling for Borrowers, GAO-09-606 (Washington, D.C.: June 29, 2009).
2
P.L. 110-289
3
The Secretary of HUD has the authority to set a minimum origination fee. For purposes of
this report, our use of the terms “HERA provisions” and “HERA changes” includes HUD’s
change to the program’s minimum origination fee.
4
Prior to HERA, some parts of Hawaii had HECM loan limits exceeding $417,000. HERA did
not change those limits. As a result of the American Recovery and Reinvestment Act of
2009, the HECM loan limit was increased in all areas of the country to $625,500 through
December 31, 2009.
500
400
300
200
100
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Age of loan in years
500
400
300
200
100
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Age of loan in years
Home equity
Source: GAO.
Note: The graphs are based on a starting house value of $300,000, with an annual 2 percent home
appreciation rate. The interest rates are assumed to be fixed at 5 percent for the forward mortgage
and 3 percent for the reverse mortgage.
There are two primary types of reverse mortgages, HECMs and proprietary
reverse mortgages. The Housing and Community Development Act of 1987
(P.L. 100-242) authorized HUD to insure reverse mortgages and
established the HECM program. According to industry officials, HECMs
account for more than 90 percent of the market for reverse mortgages.
Homeowners aged 62 or older with a significant amount of home equity
are eligible, as long as they live in the house as the principal residence, are
not delinquent on any federal debt, and live in a single-family residence. If
The volume of HECMs made annually has grown from 157 loans in fiscal
year 1990 to more than 112,000 loans in fiscal year 2008. The HECM
program has experienced substantial growth, as the number of HECMs
insured by FHA has nearly tripled since 2005 (see fig. 2).
Figure 2: Number of HECMs Insured Annually, Fiscal Years 1990 through 2008
HECMs (in thousands)
120
100
80
60
40
20
0
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
Fiscal year
Source: GAO analysis of HECM data.
80
70
60
50
40
30
20
10
0
2006 2007 2008
Fiscal year
Source: GAO analysis of FHA’s financial statements (2006 through 2008).
Finally, recent years have seen a rapid increase in the number of lenders
participating in the HECM program (see fig. 4). However, the bulk of
HECM business is concentrated among a relatively small percentage of
lenders. In fiscal year 2008, roughly 80 percent of all HECMs were
originated by fewer than 300 lenders, or about 10 percent of HECM
lenders.
2,500
2,000
1,500
1,000
500
0
2003 2004 2005 2006 2007 2008
Fiscal year
Source: GAO analysis of HECM data.
• Second, to manage its insurance risk, HUD limits the loan funds available
to the borrower by applying a “principal limit factor” to the maximum
claim amount. 5 HUD developed a principal limit factor table using
assumptions about loan termination rates—which are influenced by
borrower mortality and move-out rates—and long-term house price
appreciation rates, and indexed the table by (1) the borrower’s age and (2)
the expected interest rate—the 10-Year Treasury rate plus the lender’s
margin. The lender determines which factor to use by inputting the
borrower’s current age and the current interest rate information. The older
the borrower, the higher the loan amount; the greater the expected
interest rate of the loan, the smaller the loan amount.
5
The principal limit factor can range from 20.4 to 90 percent of the maximum claim amount.
6
Present value expresses the worth a future stream of cash inflows and outflows in terms
of an equivalent lump sum received (or paid) today. Net present value is the present value
of estimated future cash inflows minus the present value of estimated future cash outflows.
Principal _
Maximum X = Loan Up-front + Servicing = Loan funds
limit costs fee set aside
claim amount amount available
factor
FHA loan
limit Borrower’s 10 yr.
Lesser
value of or age Treasury
rate + Up-front mortgage
House lender insurance premium
value margin + origination fee
Source: GAO.
• Closing costs: HECMs also have other up-front closing costs, such as
appraisal and title search fees.
FHA’s insurance for HECMs protects borrowers and lenders in four ways.
First, lenders can provide borrowers with higher loan amounts than they
could without the insurance. Second, when the borrower is required to
repay the loan to the lender, if the proceeds from the sale of the home do
not cover the loan balance, FHA will pay the lender the difference. Third,
if the lender is unable to make payments to the borrower, FHA will assume
responsibility for making these payments. Fourth, if the loan balance
reaches 98 percent of the maximum claim amount, the lender may assign
the loan to FHA and FHA will continue making payments to the borrower
if the borrower has remaining funds in a line of credit or still is receiving
monthly payments. To cover expected insurance claims, FHA charges
borrowers insurance premiums, which go into an insurance fund. HECM
loans originated since the inception of the program through 2008 are
supported by FHA’s General Insurance and Special Risk Insurance Fund,
which includes a number of FHA mortgage insurance programs for single-
family and multifamily housing and hospitals. Pursuant to HERA, FHA
moved the HECM program and other insurance programs for single-family
housing into FHA’s Mutual Mortgage Insurance Fund.
7
FCRA was enacted as part of the Omnibus Budget Reconciliation Act of 1990 (P.L. 101-
508).
8
We surveyed a random sample of the 2,779 lenders that originated HECMs on a retail basis
in fiscal year 2008. For purposes of this report, we define retail lenders as lenders that
originate HECMs as opposed to funding HECMs originated by other lenders. For our survey
questions about HERA’s changes and other factors influencing lenders’ planned
participation, we asked lenders the following: “How, if at all, has (factor x) influenced your
institution’s likelihood to start or continue to offer HECMs on a retail basis?” Accordingly,
our results apply only to lenders’ retail HECM business. Unless otherwise noted, our
estimates have margins of error of plus or minus 10 percentage points or less at the 95
percent confidence interval. See appendix I for additional information on this survey
methodology.
HERA’s provisions on loan Moderate or great Little or no Great to moderate Don’t Not
limits and fees upward influence influence downward influence know applicable
HERA's changes to
70 28 1 1 N/A
HECM loan limits
Other factors
Implementation of the
67 29 1 2 0
HECM for Purchase program
HERA's prohibition of 9 60 29 1 0
lender-funded HECM counseling
Current availability of
9 56 22 1 12
wholesale lending partners
Note: Figure shows estimates based on GAO survey of HECM lenders. Estimates have margins of
error of plus or minus 10 percentage points or less at the 95 percent confidence interval.
Taken separately, the two HERA provisions have had differing effects on
lenders’ plans to offer HECMs. We estimate that for about 70 percent of
lenders, HERA’s increase in HECM loan limits has had an upward
Although the increase in the loan limit has generally had an upward
influence on lenders’ plans, the change to the calculation of the origination
fee has had a different effect. We estimate that changing how the fee is
calculated has had a downward influence on plans to offer HECMs for 22
percent of HECM lenders, little to no influence for 65 percent of lenders,
and an upward influence for 11 percent of lenders. Consistent with these
views, 65 percent of lenders expect the change in origination fee to have
no effect on consumer demand for HECMs. An estimated 26 percent of
lenders expect the change in the origination fee to increase consumer
demand, while only a few lenders expect the change to decrease consumer
demand.
9
See appendix II for survey results pertaining to the ARRA’s increase in the HECM loan
limit to $625,500.
10
HERA authorized a HECM for Purchase program for seniors who wish to use a HECM to
buy a new home. Unlike a traditional HECM, a HECM for purchase is made against the
value of the home to be purchased rather than against the value of a home the borrower
already owns. The HECM for purchase program allows a senior to simultaneously buy a
new home and obtain a HECM in a single transaction with a single set of closing costs,
reducing the cost to the senior.
11
HERA requires that all parties participating in the origination of HECMs be approved by
HUD. This prohibits non-HUD-approved mortgage brokers—sometimes called HECM
advisors—from offering HECMs.
• Fannie Mae officials stated that Fannie Mae bought and held more than 90
percent of HECMs in its portfolio in 2008 and was the principal secondary
market purchaser of HECM loans. However, Fannie Mae’s regulator—the
Federal Housing Finance Agency—recently required it to reduce the
mortgage assets it holds in portfolio. Fannie Mae officials told us that as a
result, they are making changes to their HECM business, which will attract
other investors to the secondary market for HECMs, in order to decrease
their share of the market. Recently, Fannie Mae lowered the price it pays
lenders for HECMs and implemented a “live pricing” system that requires
lenders to commit to the volume of HECMs they will sell to Fannie Mae. 12
We estimate that approximately 90 percent of lenders viewed secondary
market pricing requirements and the transition to live pricing as important
factors in recent margin rate increases on HECMs. Fannie officials
explained that as the price they pay lenders for HECMs falls, the margin
rate the lenders charge the consumers generally increases. 13 Some lenders
we surveyed noted that margin rate increases stemming from pricing
changes could make HECMs less attractive to borrowers because they
would not be able to obtain as much cash from their HECM. 14 Some
lenders noted that live pricing complicates their relationship with
borrowers because the interest rate can change between loan application
and closing, which may result in the senior being able to receive less
money from their HECM than originally quoted.
12
Live pricing eliminated the predetermined pricing contracts that were previously used, in
which lenders negotiated with Fannie Mae to set the price of all loans made within a
predetermined period. Instead, live pricing implemented a process whereby lenders enter
into a commitment with Fannie Mae to sell a specified volume of loans at a specified price
within 90 days. Fannie Mae officials noted that this system makes HECM pricing more
similar to the pricing system for forward mortgages in the secondary market. According to
some industry observers, some lenders are reluctant to guarantee the interest rate on a
HECM well in advance of closing because it can take several months to close a HECM and
the loan may not close before the end of the 90-day commitment period, in which case the
lender would incur a penalty.
13
Fannie Mae maintains a cap on margin rates for HECMs they purchase.
14
As previously discussed, the amount of loan funds available to the borrower depends in
part on the interest rate, which includes the lender’s margin. In general, the higher the
interest rate, the less money a borrower will have available from the HECM.
HERA Has Not Influenced In 2008, several non-HECM reverse mortgages—referred to as jumbo or
Most Lenders’ Plans to proprietary reverse mortgages—were available in the marketplace.
Offer Non-HECM Reverse Proprietary reverse mortgages offered loan limits that were greater than
the HECM loan limit. For example, Financial Freedom, a large reverse
Mortgages mortgage lender, offered a product called the Cash Account Advantage
Plan, which was not subject to the HECM loan limits, and in some cases
15
Generally, issuers can then assign the loan to HUD.
HERA’s provisions on loan Moderate or great Little or no Great to moderate Don’t Not
limits and fees upward influence influence downward influence know applicable
HERA's changes to
29 56 10 6 N/A
HECM loan limits
Other factors
Implementation of the 32 43 5 4 16
HECM for Purchase program
Current availability of
14 45 21 3 16
wholesale lending partners
HERA's prohibition of
8 61 12 4 16
lender-funded HECM counseling
Note: Figure shows estimates based on GAO survey of HECM lenders. Estimates have margins of
error of plus or minus 10 percentage points or less at the 95 percent confidence interval.
HERA Provisions Will The net effect of the HERA provisions on an individual borrower’s total
Change Up-front Costs for up-front costs depends on house value, the local loan limit prior to HERA,
Many Borrowers and the new loan limit. HECM up-front costs consist primarily of the up-
front mortgage insurance premium and the origination fee, both of which
are calculated as a proportion of the maximum claim amount. Most
borrowers are likely to see changes in origination fees due to HERA.
Generally, those with house values greater than the prior HECM loan limit
in their area will see changes in the up-front mortgage insurance
premium. 16 Borrowers fall into two categories, based on whether their
maximum claim amount changes:
• Maximum claim amount does not change: For borrowers whose houses
are valued at or less than the prior HECM loan limit in their area, the
maximum claim amount does not change. Therefore, for these borrowers,
the mortgage insurance premium (which is calculated based on the
maximum claim amount) also does not change. However, the origination
fee may change depending on the value of the house. A borrower whose
house is valued at less than $125,000 should expect up to a $500 increase
in the up-front costs due to the increase in the minimum origination fee
from $2,000 to $2,500. A borrower whose house is valued at $125,000 to
16
HERA did not change the HECM loan limits in parts of Hawaii that had loan limits higher
than $417,000 prior to HERA. As a result, borrowers in those areas whose house values are
greater than the pre-HERA loan limits will see no change in their up-front mortgage
insurance premiums due to HERA.
See appendix III for a more complete explanation of how up-front costs
will change for borrowers with different characteristics.
Up-front
Prior local mortgage
HECM loan Maximum claim insurance Origination Total up- Change in up-
Borrower limit amount premium fee front costs front costs
A $200,000 Before HERA $200,000 $4,000 $4,000 $8,000 $3,000 more
After HERA 300,000 6,000 5,000 11,000
B 275,000 Before HERA 275,000 5,500 5,500 11,000 No change
After HERA 300,000 6,000 5,000 11,000
C 290,000 Before HERA 290,000 5,800 5,800 11,600 $600 less
After HERA 300,000 6,000 5,000 11,000
D 325,000 Before HERA 300,000 6,000 6,000 12,000 $1,000 less
After HERA 300,000 6,000 5,000 11,000
Source: GAO.
17
Loan funds available for a HECM, is the loan amount after applying the principal limit
factor to the maximum claim amount and deducting the up-front costs and servicing fee
set-aside (see fig. 5).
18
For some Hawaii borrowers, the up-front mortgage insurance premium would have been
the same because their loans were capped by loan limits that did not change due to HERA.
Specifically,
• about 28 percent of borrowers would have had more loan funds available
due solely to a decrease in their up-front fees;
6
No change 27,548 0 27,554
(less than
(27%) (27%)
1%)
a
46,862 0 0 46,862
(46%) (46%)
Decrease
Source: GAO.
a
For 18,847 of these borrowers, the maximum claim amount increased. For the other 28,015
borrowers, the maximum claim amount remained the same.
$1,508 $42,560
9,761 $240,008 $64,368 (11%) (28%)
402 -402
17,303 90,760 0 (5%) (-1%)
0 14,232
6 260,875 20,292 (0%) (8%)
0 0
27,548 162,592 0 (0%) (0%)
-422 27,883
18,847 b 350,401 39,477 (-2%) (12%)
733
c -733
28,015 270,918 0 (less than
(-4%)
1%)
Increase
Decrease
No change
Source: GAO.
a
This analysis did not include refinanced HECMs, but a separate analysis of refinanced HECMs
showed similar changes in maximum claim amounts, up-front costs, and loan funds available within
the different categories of borrowers.
b
For these borrowers, the maximum claim amount increased.
c
For these borrowers, the maximum claim amount remained the same.
Borrowers May Be Increased lender margin rates stemming from HERA’s change to the
Affected by Other Factors, origination fee calculation could reduce loan funds available to borrowers.
Such as Lender Margin At loan origination, the expected interest rate HUD uses to determine the
portion of the maximum claim amount that will be made available to the
Rates and Counseling Fees borrower includes the 10-year Treasury rate plus the fixed lender margin
rate. Our survey of HECM lenders indicates that some lenders have raised
their margin rates modestly to compensate for HERA’s limitations on the
origination fee; however, we did not receive a sufficient number of
responses to reliably estimate the median increase in margin rate for the
19
Of the lenders that responded to the survey question, we estimate that 48 percent
increased margin rates for HECMs offered by their institution by at least 0.25 percentage
points. This estimate has a margin of error of within plus or minus 13 percentage points.
20
All other things being equal, an increase in margin rate causes a decrease in loan funds
available (see fig. 5). This analysis uses the 10-year Treasury rate that was available to the
borrower at loan origination.
HUD Is Taking Steps to To estimate the cost of the HECM program, HUD uses a model to project
Improve its Analysis of the the cash inflows (such as insurance premiums paid by borrowers) and
HECM Program’s Financial cash outflows (such as claim payments to lenders) for all loans over their
expected duration. HUD’s model is a computer-based spreadsheet that
Performance incorporates assumptions based on historical and projected data to
estimate the amount and timing of insurance claims, subsequent
recoveries from these claims, and premiums and fees paid by borrowers.
These assumptions include estimates of house price appreciation, interest
rates, average loan size, and the growth of unpaid loan balances. HUD
inputs its estimated cash flows into OMB’s credit subsidy calculator,
which calculates the present value of the cash flows and produces the
official credit subsidy rate for a particular loan cohort. A positive credit
subsidy rate means that the present value of the cohort’s expected cash
outflows is greater than the inflows, and a negative credit subsidy rate
means that the present value of the cohort’s expected cash inflows is
greater than the outflows. To budget for a positive subsidy an agency must
receive an appropriation. HUD also uses the cash flow model to annually
estimate the liability for loan guarantees (LLG), which represents the net
present value of future cash flows for active loans, taking into account the
prior performance of those loans. HUD estimates the LLG for individual
cohorts as well as for all cohorts combined. 21 The LLG is a useful statistic
because unusual fluctuations in the LLG can alert managers to financial
risks that require further attention.
21
The LLG does not include cash flows that have already occurred.
22
HUD OIG, Audit of the Federal Housing Administration’s Financial Statements for
Fiscal Years 2007 and 2006, 2008-FO-0002 (Washington, D.C., Nov. 8, 2007).
23
HUD OIG, Audit of the Federal Housing Administration’s Financial Statements for
Fiscal Years 2008 and 2007, 2009-FO-0002 (Washington, D.C., Nov. 7, 2008).
24
IHS Global Insight is a private company that forecasts a wide range of financial and
economic indicators.
Prior Cost Estimates HUD’s most recent estimates of two important financial indicators for the
Indicated That the HECM HECM program—the credit subsidy rate and the LLG—suggest weaker
Program Was Profitable financial performance than previously estimated, largely due to more
pessimistic house price assumptions. All other things being equal, lower
but Current Estimates house price appreciation can increase HUD’s insurance losses because it
Forecast Losses, Primarily makes it less likely that the value of the home will cover the loan balance.
Due to Revised House Analyses by HUD have found that the financial performance of the HECM
Price Assumptions program is sensitive to long-term trends in house prices. HUD officials told
us that HECM program performance is less sensitive to short-term price
25
Pursuant to congressional directives, HUD submitted reports on the HECM program in
1995, 2000, and 2003 (P.L. 100-242 and P.L. 106-569). These reports included actuarial
reviews of the program.
26
12 U.S.C. Sec. 1708(a)(4). HECMs originated in fiscal year 2009 and beyond will be
accounted for under the MMI Fund. Loans originated in or before fiscal year 2008 will be
accounted for under the GI/SRI Fund.
27
Currently, HUD includes HECMs in its credit subsidy re-estimates for the GI/SRI Fund as
a whole.
HUD has made credit subsidy estimates for HECM cohorts from 2006
forward. Because the HECM program was relatively small prior to 2006,
HUD did not produce separate subsidy estimates for the HECM program
but included HECMs in its estimates of subsidy costs for the GI/SRI Fund
as a whole. For the 2006 through 2009 HECM cohorts, HUD estimated
negative subsidy rates ranging from - 2.82 percent in 2007 to -1.37 percent
in 2009 (see fig. 10). However, for the 2010 cohort, HUD estimated a
positive subsidy rate of 2.66 percent. Because HUD is expecting to insure
about $30 billion in HECMs in 2010, this rate corresponds to a subsidy cost
of $798 million. As required by the Federal Credit Reform Act, the
President’s budget for fiscal year 2010 includes a request for this amount. 28
28
On July 17, 2009, the House Committee on Appropriations approved a fiscal year 2010
appropriations bill for HUD. The bill contained a provision directing HUD to adjust, as
necessary, the principal limit factor for new loans to ensure that the program operates
without a subsidy. The provision also prohibited HUD from reducing a borrower’s principal
limit factor below 60 percent. If enacted, this provision would eliminate the need for the
$798 million budget request but would also reduce the loan funds available to some
borrowers.
1
Yes
0 Subsidy required?
No
-1
-1.37%
-2 -1.74% -1.90%
-3 -2.82%
2006 2007 2008 2009 2010
Cohort
Source: GAO analysis of OMB federal credit supplement data (2006 through 2010).
HUD officials told us that the positive subsidy rate for fiscal year 2010
largely was due to incorporating more conservative assumptions about
long-term house price trends than had been used for prior cohorts. For
budgeting purposes, the Administration decided to use more modest
appreciation rates than the private sector forecasts HUD typically uses.
Specifically, the house price appreciation rates used were 0.5 percent
greater than the forecasted inflation rates. HUD officials told us that if
they had used IHS Global Insight projections to develop the fiscal year
2010 credit subsidy estimate, there would be no need for an appropriation
because the credit subsidy rate would be negative.
HUD also has estimated the LLG for the HECM program since 2006. As
shown in figure 11, HUD’s original LLG estimates grew substantially from
2007 to 2008, increasing from $326 million to $1.52 billion. According to
FHA’s financial statements for fiscal years 2007 and 2008, the increase was
primarily due to the lower house price appreciation projections used in the
2008 analysis. 29 The report noted that lower appreciation rates result in
29
HUD, FHA Annual Management Report Fiscal Year 2007 (Washington, D.C., 2007) and
FHA Annual Management Report Fiscal Year 2008 (Washington, D.C., November 2008).
Figure 11: Loan Liability Guarantee for the HECM Program, Fiscal Years 2006
through 2008
1,400
1,200
1,000
800
600
400
200
0
2006 2007 2008
Fiscal year
Source: GAO analysis of FHA financial statements (2006 through 2008).
In September 2008, HUD analyzed the sensitivity of the 2008 LLG estimate
for the HECM program as a whole to different assumptions, including
alternative house price scenarios. HUD examined the impact of house
price appreciation that was 10 percent higher and 10 percent lower than
the baseline assumptions from IHS Global Insight for fiscal years 2009
through 2013. (For example, for a baseline assumption of 4 percent house
price appreciation, the lower and higher scenarios would have been 3.6
percent and 4.4 percent, respectively.) HUD estimated that the more
pessimistic assumption increased the LLG from $1.52 billion to $1.78
billion, while the more optimistic assumption reduced the LLG to $1.27
billion.
Higher loan limits enacted under HERA and ARRA may make HUD’s
approach less conservative by reducing the proportion of loans for which
the property value exceeds the maximum claim amount. This scenario is
especially likely in locations that previously had relatively low local loan
limits (reflecting their lower home values) but are now subject to the
higher national limit. To illustrate, consider a 65-year-old HECM borrower
with a $400,000 home whose loan limit prior to HERA was $250,000 (see
fig. 12). In this scenario, the maximum claim amount would be the same as
the loan limit because the maximum claim amount is defined as the lesser
of the loan limit or the home value. However, if the loan limit for the same
borrower is increased to the HERA-authorized level of $417,000, the
maximum claim amount is the same as the home value ($400,000). As
figure 12 shows, when a borrower’s maximum claim amount is capped by
the loan limit, the maximum claim amount can be substantially lower than
the value of the home. All other things being equal, the potential for losses
is low in this scenario because the projected loan balance is likely to
remain less than the projected home value after the lender assigns the loan
to HUD. 30 In contrast, when the maximum claim amount is capped by the
home’s value, the difference between the projected loan balance and the
projected home value is smaller. The potential for losses is higher with
such a loan because the projected loan balance is more likely to exceed
the projected home value. As also shown in figure 12, when this effect is
combined with declining home prices, the potential for losses increases.
30
As previously noted, lenders may assign the loan to HUD when the loan balance reaches
98 percent of the maximum claim amount. At that point, HUD takes over servicing of the
loan and is responsible for covering any losses.
Pre-HERA Post-HERA
1,200 1,200
1,000 1,000
800 800
ce
lan
600 e 600 e ba
alu alu
ty v ty v an
per e pe
r
Lo
Pro lanc Pro $400,000
400 n ba 400
a max claim amount
Lo Loan
$250,000
assigned
200 max claim amount 200 to HUD
Loan
assigned
to HUD
0 0
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Age of loan in years Age of loan in years
Source: GAO.
Note: The trend line for property value assumes house price appreciation of 4 percent annually—this
is the rate HUD assumes in its pricing of the HECM loan product. The initial loan amount is based on
a $400,000 home, as well as an expected interest rate of 5.43 percent and a borrower age of 65,
which corresponds to a principal limit factor of 64.9 percent. The trend line for loan balance assumes
that the borrower immediately drew down 100 percent of the available loan funds.
Studies by HUD and others have noted that HECM loans for which the
home value exceeds the maximum claim amount have a positive impact on
the program’s financial performance but also have noted the potential
negative impact of raising the loan limit. When the HECM program started
in 1990, HUD developed a statistical model to estimate borrower payments
and insurance risk. HUD’s technical explanation of the model
acknowledges that future expected losses are smaller for HECMs with a
31
HUD, The FHA Home Equity Conversion Mortgage Insurance Demonstration: A Model
to Calculate Borrower Payments and Insurance Risk (Washington, D.C., October 1990).
32
HUD, Evaluation of the Home Equity Conversion Mortgage Insurance Demonstration:
Report to Congress. (Washington, D.C., 1995); Evaluation Report of FHA’s Home Equity
Conversion Mortgage Insurance Demonstration. Prepared by Abt. Assoc. (Washington,
D.C., Mar. 31, 2000); Refinancing Premium, National Loan Limit, and Long-Term Care
Premium Waiver For FHA’s HECM Program (Final Report). Prepared by Abt. Assoc.
(Washington, D.C., May 2003).
33
CBO, Cost Estimate: FHA Modernization Act of 2007 (Washington, D.C., Oct. 12, 2007).
90 18
27 25 24 25
32 29 31 31 31
40 38 39 34 36
80 39 41 42
48 47
70
60
50
82
40 73 76 75
71 75
68 69 69 69
30 60 61 62 61 66 64 59
52 53 58
20
10
0
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Fiscal year
Note: Fiscal year 2009 data include HECMs insured from October 1, 2008, through January 31, 2009.
Mathew J. Scirè
Director, Financial Markets
and Community Investment
Methodology
Our objectives were to examine (1) how the Housing and Economic
Recovery Act of 2008 (HERA) changes to the Home Equity Conversion
Mortgage (HECM) program and other factors have affected HECM lenders’
planned participation in the reverse mortgage market, (2) the extent to
which HERA’s changes to HECM origination fees and loan limits will
affect costs to borrowers and the loan amounts available to them, and (3)
Department of Housing and Urban Development’s (HUD) actions to
evaluate the financial performance of the HECM program, including the
potential impact of loan limit and house price changes.
thus would not influence overall estimates very much. Responses from the
smallest lenders stratum were used only as case study examples in our
analysis.
Fiscal
year 2008 Original Total Ineligible Usable Non- Response
a b
Stratum Population HECMs sample responses responses responses responses e rate
Very small (1-9
HECM loans in
2008)c 1,837 5,549 30 10 2 8 20 .80 33.3
Small (10 to 59 669 16,132 170 83 3 80 87 .96 48.8
loans in 2008)
Medium (60 to 222 27,247 113 66 13 53 47 .80 58.4
299 loans in
2008)
Large (300 or 51 63,087 51 41 2 39 10 .95 80.8
more loans in
2008)
Totalc 2,779 112,015 364 200 20 180 164 .91 56.9%c
Source: GAO.
a
e = estimated eligibility rate of nonresponding sample cases whose eligibility is unknown.
b
Response rate is unweighted, as defined by AAPOR RR3.
c
“Very small” stratum not included in overall response rate calculations.
Before the survey, in early March 2009, NRMLA sent letters to those
lenders in our sample who were also members in that organization,
endorsing our survey and encouraging response. In March 2009, we sent e-
mails with links to our Web questionnaire and unique login information to
each member of our sample with valid e-mail addresses. For sampled
companies for which we were unable to obtain working e-mail addresses,
we mailed paper versions of the questionnaires. Nonresponding lenders
were sent additional e-mails or copies of questionnaires from March
through May. We also made telephone calls in April to nonrespondents
encouraging them to respond. Our survey closed in early May 2009. We
1
We used the response rate definition “RR3,” as defined by the American Association for
Public Opinion Research in “Standard Definitions: Final Dispositions of Case Codes and
Outcome Rates for Surveys,”
http://www.aapor.org/uploads/Standard_Definitions_07_08_Final.pdf, p. 35.
2
We grouped the sample lenders into two groups defined by when they were first approved
to offer HECM loans (before 2000 and 2000 or later), and found no association between the
year category and whether the lender responded. In addition, we compared sampled
lenders in the 5 states with the most HECM loans to the rest of the sampled lenders in the
remaining states and also found no association with whether the lender responded.
Second, we applied the HERA changes to HUD loan-level data for HECMs
that borrowers obtained in calendar year 2007. We compared the results to
the actual up-front costs and loan funds available for these borrowers. To
perform this analysis, we obtained data from HUD’s Single-family Data
Warehouse. We assessed the reliability of these data by (1) reviewing
existing information about the data and the system that produced them,
(2) interviewing HUD officials knowledgeable about the data, and (3)
performing electronic testing of required data elements. We determined
that the data we used were sufficiently reliable for the purposes of this
report. As shown in table 3, the universe of 2007 HECMs used in our
analysis included 101,480 loans. 4 We applied the $417,000 national loan
limit and HERA’s changes to the origination fee calculation to the 2007
HECMs. For each borrower, we calculated the new maximum claim
amount, origination fee, up-front mortgages insurance premium, and loan
funds available under the HERA rules and compared our results to the
actual 2007 values. We summarized our results by calculating the average
changes in these amounts.
3
We used $625,500 as a maximum because ARRA resulted in an increase in the HECM loan
limit to that amount.
4
The 101,480 loans did not include 6,664 HECMs obtained in 2007 to refinance an existing
HECM. We analyzed refinanced HECMs separately because the calculation for the
mortgage insurance premium is different for these loans than for new HECMs. Our results
for the analysis of refinanced HECMs were similar to those for new HECMs.
Table 3: Universe of 2007 HECMs by Home Value and Maximum Claim Amount
Category
5
As discussed in the body of this report, our survey of HECM lenders indicated that some
lenders have raised their margin rates modestly to compensate for HERA’s limitations on
the origination fee. We did not receive a sufficient number of responses to reliably estimate
the average increase in margin rate for the population. Of the lenders that responded to the
survey question, we estimate that 48 percent increased margin rates for HECMs offered by
their institution by at least 0.25 percentage points. This estimate has a margin of error of
within plus or minus 13 percentage points.
Figure 14: Influence of ARRA’s Increase to Loan Limits on Lenders’ Plans to Offer Reverse Mortgages
Note: Figure shows estimates based on GAO survey of HECM lenders. Estimates have margins of
error of plus or minus 10 percentage points or less at the 95 percent confidence interval.
Figure 15: Influence of ARRA’s Increase to Loan Limits on Consumer Demand for HECMs
Note: Figure shows estimates based on GAO survey of HECM lenders. Estimates have margins of
error of plus or minus 10 percentage points or less at the 95 percent confidence interval.
There are two up-front costs. The first—the up-front mortgage insurance
premium—is 2 percent of the maximum claim amount. The second—the
origination fee—was calculated before HERA as 2 percent of the
maximum claim amount with a minimum fee of $2,000. HERA changed the
calculation of the origination fee to 2 percent of the first $200,000 of the
maximum claim amount plus 1 percent of the maximum claim amount
over $200,000, with a maximum fee of $6,000. In implementing HERA,
HUD also increased the minimum origination fee by $500 to $2,500.
Figure 16 shows the results for borrowers who have home values lower
than the previous loan limit. The maximum claim amount is not affected
by HERA’s change in loan limit. Therefore, for these borrowers, changes in
up-front costs derive only from changes in the origination fee.
1
HERA did not change the HECM loan limits in parts of Hawaii that had loan limits higher
than $417,000 prior to HERA. As a result, borrowers in those areas whose house values are
greater than the pre-HERA loan limits will see no change in their up-front mortgage
insurance premiums due to HERA.
Figure 16: Changes to Up-front Costs for Borrowers Not Affected by HERA’s
Change in Loan Limit
$125,000 to $200,000 0 $0
Source: GAO.
Figure 17 shows the results of the calculation for borrowers who were
affected by HERA’s increase in loan limit. These borrowers would pay up-
front mortgage insurance premiums and origination fees based on a higher
maximum claim amount. However, depending on the maximum claim
amount, the origination fee may have decreased rather than increased. The
net change in up-front costs for this grouping is therefore indeterminable
without knowing the old and new maximum claim amounts.
Figure 17: Changes to Up-front Costs for Borrowers Affected by HERA’s Change in Loan Limit
Note: New maximum claim amounts and old maximum claim amounts less than $125,000 are not
valid because no local loan limits were less than $125,000.
Acknowledgments
(250421)
Page 48 GAO-09-836 Reverse Mortgages
GAO’s Mission The Government Accountability Office, the audit, evaluation, and
investigative arm of Congress, exists to support Congress in meeting its
constitutional responsibilities and to help improve the performance and
accountability of the federal government for the American people. GAO
examines the use of public funds; evaluates federal programs and policies;
and provides analyses, recommendations, and other assistance to help
Congress make informed oversight, policy, and funding decisions. GAO’s
commitment to good government is reflected in its core values of
accountability, integrity, and reliability.
The fastest and easiest way to obtain copies of GAO documents at no cost
Obtaining Copies of is through GAO’s Web site (www.gao.gov). Each weekday afternoon, GAO
GAO Reports and posts on its Web site newly released reports, testimony, and
correspondence. To have GAO e-mail you a list of newly posted products,
Testimony go to www.gao.gov and select “E-mail Updates.”
Order by Phone The price of each GAO publication reflects GAO’s actual cost of
production and distribution and depends on the number of pages in the
publication and whether the publication is printed in color or black and
white. Pricing and ordering information is posted on GAO’s Web site,
http://www.gao.gov/ordering.htm.
Place orders by calling (202) 512-6000, toll free (866) 801-7077, or
TDD (202) 512-2537.
Orders may be paid for using American Express, Discover Card,
MasterCard, Visa, check, or money order. Call for additional information.
Contact:
To Report Fraud,
Waste, and Abuse in Web site: www.gao.gov/fraudnet/fraudnet.htm
E-mail: fraudnet@gao.gov
Federal Programs Automated answering system: (800) 424-5454 or (202) 512-7470